In the semiconductor business field, it is a target to be pursued at all times that the yield of semiconductor products is enhanced and the quality level is kept high. Recently, semiconductor manufacturing apparatuses are particularly closed up under the situation that the design quality of semiconductor devices is relatively stable, and it has been strongly required to enhance the quality of semiconductor devices and keep high quality by semiconductor manufacturing apparatuses. Furthermore, for the purpose of optimization of the overall semiconductor manufacturing process, it has been strongly required to enhance the operation rate of the semiconductor manufacturing apparatus such as to shorten the start-up time of the semiconductor manufacturing apparatus and the correspondence time to occurrence of a failure.
Furthermore, the semiconductor manufacturing apparatus is constructed by a large number of parts, and secular distortion is unavoidable in many parts. Therefore, in order to operate the semiconductor manufacturing apparatus with high quality without inducing any failure, it is effective to check visually and periodically whether the parts constituting the semiconductor manufacturing apparatus operates normally or not. However, the visual check is unrealistic in consideration of complexity of the semiconductor manufacturing apparatus and the multitude of the number of parts.
Therefore, in a conventional substrate processing system, there has been adopted a method of converting to an electrical signal an output of a sensor or the like which is required to be periodically checked, taking the electrical signal into the main controller of the semiconductor manufacturing apparatus and comparing it with upper and lower limit values preset in the main controller to thereby judge whether each part normally operates.